Abstract:In order to solve the problem in the structured lane recognition algorithm which includes many constraints and complicated calculation, this paper presents a fast lane detection algorithm based on morphology. Firstly, this work gets pre-process in road image by handled Median filter. Sobel operator is to enhance edge. A characteristic histogram threshold method is adopted to segment images. And region of interest is disposed with the detection results. Secondly, by constructing morphological structure elements with the feature of lane start to extract the lane line. Finally, Hough transform have been fitting with the best line of lane. The contrast experiment shows that algorithm provides a high real-time, less calculation and simple and effective approach to structured lane.